• Open Access

Cervical Myelomalacia in a Lamb (Ovis aries) and an Alpaca (Vicugna pacos) after Attempted Intramuscular Injection


Corresponding author:

Amy L. Johnson, New Bolton Center, 382 West Street Road, Kennett Square, PA 19348; e-mail: amyjohn@vet.upenn.edu


glial fibrillary acidic protein


Luxol fast blue-periodic acid Schiff

A 4.5-kg, 2-day-old Scottish Blackface ram lamb was presented to its regular attending veterinarian for presumptive aspiration pneumonia. The veterinarian prescribed florfenicol at 20 mg/kg IM, and 0.3 mL of florfenicol solution was prepared for planned administration into the cervical musculature using a 20 gauge 1-inch needle. The lamb was restrained and a veterinary student inserted the needle into the right cervical musculature, aspirated to ensure no blood appeared in the hub of the needle, and then injected the solution. The lamb reportedly struggled during the injection. Immediately after the injection, the lamb developed muscle tremors in its pelvic limbs, a non-weight-bearing lameness in its right thoracic limb, and cervical scoliosis. Within a minute the lamb collapsed, making no attempts to stand.

Treatment consisted of 0.1 mg/kg dexamethasone IV and temporarily splinting the right thoracic limb in extension. The lamb remained unable to stand or ambulate, but was responsive and drank from a bottle when offered. No overtly painful areas were detected, although the lamb appeared lethargic and occasionally vocalized. Dexamethasone administration was repeated 1 day after injection (day 2). No changes in the lamb's neurologic signs were seen during the first 3 days post injection.

On day 4, the lamb was presented to the university hospital for a complete neurologic examination. The lamb was alert and responsive with a strong suckle and no cranial nerve abnormalities, but was nonambulatory and unable to stand without support. The right thoracic limb was flaccid with no volitional motor activity seen, whereas the other 3 limbs had normal muscle tone and motor activity with paresis and proprioceptive deficits. Spinal segmental reflexes, including withdrawal, were severely decreased to absent in the right thoracic limb, but present and normal or increased in the other 3 limbs. The lamb was severely hypalgesic-to-analgesic over the right side of the neck from the level of the caudal aspect of the axis (C2) to the scapular spine, but normal responses to noxious stimuli were evident when the right thoracic limb was stimulated from the elbow distally. Normal nociception was also observed when the left side of the neck was stimulated. Cervical scoliosis with the convex side of the curve on the right and the concave side on the left was present. These abnormalities were consistent with an extensive cervical spinal cord lesion involving the right dorsal grey column from the cranial cervical spinal segments to the cervical intumescence, as well as damage to white matter bilaterally in the cervical region and to the right ventral grey column in the cervical intumescence.

Supportive care was continued for several more days. The lamb remained alert and responsive with a good appetite, but showed no improvement in neurologic status. On day 8, the lamb was euthanized.

Postmortem examination revealed lateral deviation of the cervical vertebral column (convex to the right) with mild atrophy of the right cervical musculature. There was focally extensive epidural hemorrhage in the right lateral aspect of the vertebral canal from the caudal aspect of the 2nd cervical vertebra to the cranial aspect of the 5th cervical vertebra. The vertebral sinuses were prominent and distended with clotted blood. A 1 × 0.5 cm rectangular focus of subdural hemorrhage was present on the dorsal aspect of the spinal cord at the level of the 3rd nerve root (Fig 1A), and there was mild discoloration of the spinal cord within the 3rd through the 5th cervical vertebrae. Histologic examination was performed using Luxol fast blue-periodic acid Schiff (LFB-PAS) histochemical stain to highlight myelin, glial fibrillary acidic protein (GFAP) immunohistochemical stain to highlight astrocytes, and Perl's iron stain to highlight hemosiderin. Histologic examination of the 3rd through 7th cervical spinal cord segments showed a well-demarcated, unilateral focus of liquefactive necrosis, consistent with an injection track, which at the level of the 3rd cervical segment extended from the right dorsolateral periphery of the spinal cord through adjacent white matter into the dorsal horn (Fig 1B, C), and primarily into the dorsal horns of adjacent segments. White matter degeneration within the opposite (left) lateral funiculus was identified only within the 3rd cervical spinal segment. Within the necrotic track, were numerous myelinophages, hemorrhage, mild erythrophagocytosis and hemosiderosis, minimal neutrophilic infiltrates, and rare gemistocytic astrocytes (Fig 1D–F). Hematoidin and neovascularization were not observed. Wallerian degeneration was observed in the immediately adjacent parenchyma as well as in the left lateral funiculus, characterized by dilated and fragmented myelin sheaths, swollen axon fibers (spheroids), and myelinophages with scattered chromatolytic neurons (Fig 1G). In some regions, a pale, LFB-PAS negative, eosinophilic, homogenous paucicellular fluid was observed; this material could not be identified. Histologic examination of the right cervical musculature revealed focal myocyte degeneration and necrosis as well as small angular myocytes consistent with denervation atrophy.

Figure 1.

Case 1. Gross photograph (panel A) that shows right lateral scoliosis of the cervical vertebral column and a 1 × 0.5 cm focus of hemorrhage located on the dorsal aspect of the spinal cord at the level of the 3rd cervical nerve root (white arrow). Low (10× mag., bar = 500 μm) magnification photomicrographs of a Luxol fast blue-periodic acid Schiff (LFB-PAS) histochemical stained section that highlights myelin (panel B), and a glial fibrillary acidic protein (GFAP) immunohistochemical stained section (panel C) that highlights astrocytes and astrocytic processes of corresponding spinal segments reveal a well-demarcated unilateral focus of liquefactive necrosis that extends through both white and gray matter centered around the right dorsal horn (panel B, arrow heads; panel C, asterisk). Wallerian degeneration is located within the parenchyma immediately adjacent to the track as well as in the left lateral funiculus of this spinal segment (panels B, C, paired arrows). Photomicrographs of LFB-PAS (400× mag., bar = 20 μm) and GFAP (200× mag., bar = 50 μm) stained sections reveal fragmentation and loss of myelin sheaths with scattered swollen axon fibers, ie, spheroids (panel D, S), myelinophages (panel D, arrows), and few “reactive” gemistocytic astrocytes (panel E, arrows). Within areas of hemorrhage, a Perl's iron stained section (400× mag., bar = 20 μm) reveals hemosiderin-laden macrophages (panel F, arrows), indicating erythrophagocytosis. A LFB-PAS stained photomicrograph (200× mag., bar = 50 μm) highlights chromatolytic neurons characterized by loss of Nissl substance and karyolysis (panel G, arrowhead) admixed with viable neurons containing prominent Nissl substance and karyon (panel G, arrows).

The second case, a 32-kg 16-month-old female alpaca, was presented with similar history and signs. Four days before presentation, the alpaca reportedly was administered IM doses of 2 different vaccines and 1 injectable anthelmintic medication by a veterinarian. According to the owner of the animal, the vaccines were a combination clostridial product (C. perfringens types C and D plus C. tetani toxoids) and rabies, and the anthelmintic was ivermectin. Within 30 seconds of receiving the 3rd injection, the alpaca reportedly became unable to bear weight on the thoracic limbs and then over a short period of time lost the ability to support weight with the pelvic limbs. Seven other alpacas on the farm were administered the same 3 products by the veterinarian on the same day and remain unaffected. The affected alpaca was initially treated with flunixin meglumine IV, dexamethasone IV, dilute dimethyl sulfoxide IV, and omeprazole PO (dosages not reported).

Upon presentation, the alpaca was lethargic, recumbent, and tachycardic (100 beats/minute). Neurologic evaluation revealed nonambulatory tetraparesis. Normal nociception was present when all 4 limbs were stimulated; however, the withdrawal reflex was absent in the right thoracic limb. Spinal segmental reflexes were increased in both pelvic limbs and normal in the left thoracic limb. On the basis of these findings, a cervical spinal cord lesion that was worse on the right side and involved grey matter of the cervical intumescence was suspected. A cerebrospinal fluid sample was collected via lumbosacral approach for analysis. The cerebrospinal fluid contained 0 nucleated cells/μL (normal < 5 nucleated cells/μL), 324 red blood cells/μL (normal 0 RBCs/uL), and a total protein level of 161 mg/dL (normal < 60 mg/dL). This albuminocytologic dissociation was considered consistent with spinal cord trauma or compression. The alpaca was given a poor prognosis based on prolonged recumbency with no evidence of improvement and euthanasia was elected.

Postmortem examination was performed within 1 hour after euthanasia. When the neck of the alpaca was shaved, a pinpoint red lesion consistent with an injection site was located on the right side laterally over the caudal part of the cervical vertebral column. Two additional pinpoint lesions, 1 on the right and 1 on the left, were evident dorsal to the vertebral column. A focal area of hemorrhage was present within the subcutaneous musculature at the site overlying the vertebral column. Histologically, this area demonstrated moderate, subacute, locally extensive muscular necrosis with hemorrhage, and lymphohistiocytic infiltration; these findings were considered consistent with an injection site. Although there was no gross or histologic evidence of communication between this area and the cervical spinal cord, histology of the spinal cord at this level revealed moderately severe acute unilateral myelomalacia of white and grey matter with ballooning of myelin sheaths, astrocytic gliosis, and frequent spheroids.

As demonstrated by these cases, attempted intramuscular administration of medication, generally considered a simple and benign procedure, might occasionally result in life-threatening complications. The true incidence of intramuscular injection complications in veterinary patients remains unknown, but in human medicine estimates range from 0.4 to 28% depending on patient population and drug administered.[1, 2] These complications range from mild and self-limiting pain to life-threatening sepsis or permanent neurologic deficits. In the 2 patients reported herein, intended intramuscular administration of common drugs resulted in immediate onset of neurologic signs that eventually necessitated euthanasia. In case 1, additional special stains confirmed that the age of the observed lesion corresponded with the date of injection. Erythrophagocytosis and hemosiderosis were confirmed using H&E and Perl's iron histochemical stains. Neither neovascularization, which typically begins 7–10 days postinjury, nor astrocytic sclerosis, which takes weeks to months to develop, was present, suggesting a subacute lesion of 5–10 days duration.[3] The clinical history and the discrete unilateral “track” lesion are strong evidence for needle, drug penetration, or both of the spinal cord. In case 2, an injection site was immediately adjacent to the affected area of spinal cord, but no direct communication was seen. The mechanisms of spinal cord damage, in this case, remain unclear, but could include direct damage from the needle or administered drug, inadvertent intravascular injection of the drug, or diffusion of drug from a paravertebral site into the vertebral canal. The cervical spinal cord receives its blood supply via spinal branches of the vertebral artery; inadvertent injection into these vessels or the accompanying veins could potentially lead to localized spinal cord infarction. Proposed mechanisms for infarction include intra-arterial injection of a particulate drug leading to embolization, arterial perforation causing dissection, thrombosis, or both, and needle-induced vasospasm.[4] Intra-arterial injection is generally considered the most likely mechanism, and a recent case report documents multiple arterial thrombi presumed secondary to accidental intra-aortic injection of ethanol as the cause of spinal cord infarction in a calf.[5] Although vascular lesions were not identified in these cases, they could have been missed.

This report is the first to document acute spinal cord injury as a complication of intramuscular injection; surprisingly, few cases of iatrogenic injection complications are reported in the veterinary literature. In fact, the only report of iatrogenic spinal cord damage in farm animal patients after intramuscular injection documents an outbreak of caudal paresis in 56/610 beef cattle, with signs occurring 11–34 days after injection of an adjuvanted bacterin into loin muscle.[6] These animals developed large pyogranulomatous inflammatory masses at the sites of vaccination that spread through adjacent intervertebral foramina into the vertebral canal, compressing the lumbar spinal cord and nerve roots.

Human medical literature contains a limited number of reports of acute spinal cord injury after intramuscular or epidural injections. The lesions are generally described as infarcts and vascular etiologies are proposed. For example, several cases of transverse myelopathy around the level of the 10th thoracic spinal cord segment were reported after injection of a viscid antibiotic solution into the gluteal musculature.[7-10] Authors hypothesized that inadvertent intra-arterial injection of the medication resulted in occlusive vascular disease and retrograde delivery of the medication into vessels supplying the spinal cord, causing a transverse myelopathy. More recently, patients with brain and spinal cord infarctions after transforaminal cervical epidural steroid injections have been described; the proposed mechanism again is inadvertent intra-arterial injection.[4]

Although injection complications can clearly be life-threatening, surprisingly little information is available regarding ideal techniques to avoid these complications in veterinary species. In farm animal medicine, most research into injection techniques has been focused on improving meat quality by reducing tissue damage and tenderness complications created by injection-site damage in prime cuts of meats. As muscles of the hindquarters are more valuable than cervical musculature, emphasis has been placed on training producers to use the neck region for all pharmaceutical products.[11] Veterinary textbooks also recommend intramuscular injection in the neck region, and frequently caution that inappropriate injection technique or irritating medications in the hindquarters may cause sciatic nerve damage.[12-14] For these reasons, many farm animal veterinarians are accustomed to cervical intramuscular injections, which are suitable for adult cattle, but in the authors' opinion contraindicated in camelids and small ruminants. It is obvious that a blanket recommendation for injection technique is inappropriate for veterinary medicine, which is unique in that practitioners are frequently presented with patients of many different sizes and anatomic conformations. Rather than recommend a particular injection technique, this report highlights the need for careful consideration of potential complications before injection of any medication via any route. As both small ruminants and camelids have relatively thin cervical musculature compared to other large animal species, intramuscular injection in the cervical region poses a higher degree of risk.

In summary, this report documents a previously unrecognized complication of intramuscular injection in small livestock. Neurologic signs in both cases were similar and reflected asymmetric spinal cord damage extending over several spinal segments ipsilateral and at the level of the injection site. Because this complication has not been previously reported, it may be rare or simply underrecognized and underreported in veterinary patients. Because the severity of signs, in both cases, necessitated euthanasia, practitioners, educators, and pathologists should be cognizant of this potential complication to assist in preventing its occurrence as well as to consider it as a differential diagnosis for any case of acute spinal cord disease, particularly those with a history of receiving an injectable product before onset of signs. Only with adequate recognition of this complication can its incidence be determined, and this information may assist in the development of future injection technique recommendations.


Conflict of Interest: Authors disclose no conflict of interest.